DESIGN AND CONSTRUCTION OF A MICROWAVE PLASMA ION … · February 2011, 56 pages This thesis is...
Transcript of DESIGN AND CONSTRUCTION OF A MICROWAVE PLASMA ION … · February 2011, 56 pages This thesis is...
DESIGN AND CONSTRUCTION OF A MICROWAVE PLASMA ION SOURCE
A THESIS SUBMITTED TOTHE GRADUATE SCHOOL OF NATURAL AND APPLIED SCIENCES
OFMIDDLE EAST TECHNICAL UNIVERSITY
BY
KAMIL CINAR
IN PARTIAL FULFILLMENT OF THE REQUIREMENTSFOR
THE DEGREE OF MASTER OF SCIENCEIN
PHYSICS
FEBRUARY 2011
Approval of the thesis:
DESIGN AND CONSTRUCTION OF A MICROWAVE PLASMA ION SOURCE
submitted by KAMIL CINAR in partial fulfillment of the requirements for the degree ofMaster of Science in Physics Department, Middle East Technical University by,
Prof. Dr. Canan OzgenDean, Graduate School of Natural and Applied Sciences
Prof. Dr. Sinan BilikmenHead of Department, Physics
Prof. Dr. Sinan BilikmenSupervisor, Physics Dept., METU
Examining Committee Members:
Assoc. Prof. Dr. Serhat CakırPhysics Department, METU
Prof. Dr. Sinan BilikmenPhysics Department, METU
Assoc. Prof. Dr. Ismail RafatovPhysics Department, METU
Dr. Burak YedierlerPhysics Department, METU
Dr. Ali AlacakırResearch and Development Department, SANAEM
Date:
I hereby declare that all information in this document has been obtained and presentedin accordance with academic rules and ethical conduct. I also declare that, as requiredby these rules and conduct, I have fully cited and referenced all material and results thatare not original to this work.
Name, Last Name: KAMIL CINAR
Signature :
iii
ABSTRACT
DESIGN AND CONSTRUCTION OF A MICROWAVE PLASMA ION SOURCE
Cınar, Kamil
M.Sc., Department of Physics
Supervisor : Prof. Dr. Sinan Bilikmen
February 2011, 56 pages
This thesis is about the designing and constructing a microwave ion source. The ions are
generated in a thermal and dense hydrogen plasma by microwave induction. The plasma is
generated by using a microwave source with a frequency of 2.45 GHz and a power of 700 W.
The generated microwave is pulsing with a frequency of 50 Hz. The designed and constructed
microwave system generates hydrogen plasma in a pyrex plasma chamber. Moreover, an ion
extraction unit is designed and constructed in order to extract the ions from the generated
hydrogen plasma. The ion beam extraction is achieved and ion currents are measured. The
plasma parameters are determined by a double Langmuir probe and the ion current is mea-
sured by a Faraday cup. The designed ion extraction unit is simulated by using the dimensions
of the designed and constructed ion extraction unit in order to trace out the trajectories of the
extracted ions.
Keywords: Ion Source, Microwave Plasma, Dense Plasma, Ion Generation, Ion Extraction
iv
OZ
MIKRODALGA PLAZMA IYON KAYNAGI TASARIMI VE URETIMI
Cınar, Kamil
Yuksek Lisans, Fizik Bolumu
Tez Yoneticisi : Prof. Dr. Sinan Bilikmen
Subat 2011, 56 sayfa
Bu tezin konusu mikrodalga iyon kaynagı tasarımı ve uretimidir. Iyonlar, yogun ve ter-
mal hidrojen plazması icinde uretilmistir. Plazma, 2.45 GHz frekanslı ve 700 Watt cıkıs
gucunde mikrodalga kaynagı kullanılarak olusturulmustur. Uretilmis mikrodalga kaynagı, 50
Hz’lik frekans atımlı mikrodalga olusturmaktadır. Tasarlanan ve uretilen mikrodalga sistemi,
payreks plazma kabı icinde hidrojen plazması uretmistir. Ayrıca iyonları olusturulan hidro-
jen plazmasından cıkarmak icin iyon cıkarma unitesi tasarlanıp uretilmistir. Iyon cıkarma
basarılmıstır. Plazma parametreleri cift Langmuir sondası ile, iyon akımı ise Faraday kabı ile
olculmustur. Tasarlanan iyon cıkarma unitesi boyutları kullanılarak iyon yorungeleri simule
edilmistir.
Anahtar Kelimeler: Iyon Kaynakları, Mikrodalga Plazma, Yogun Plazma, Iyon Uretimi, Iyon
Cıkartımı
v
To my parents
vi
ACKNOWLEDGMENTS
I express my sincere appreciation to Prof. Dr. Sinan Bilikmen who provided the opportunity
of working on the present subject. I respectfully acknowledge Dr. Ali Alacakır for this contin-
uous support on not only proffessional issues but also on personal matters. His unique effort
on creating a very productive working environment will always be appreciated. Moreover, my
special thanks goes to Dr. Erdal Recepogulu and Dr. Hande Karadeniz for their invaluable
support.
vii
TABLE OF CONTENTS
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
OZ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
CHAPTERS
1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 ION PRODUCTION VIA PLASMA . . . . . . . . . . . . . . . . . . . . . . 4
2.1 PLASMA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2 DC DISCHARGE . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 MICROWAVE DISCHARGE . . . . . . . . . . . . . . . . . . . . . 12
3 EXTRACTION OF IONS FROM PLASMA . . . . . . . . . . . . . . . . . . 18
3.1 PLASMA (SPACE) POTENTIAL . . . . . . . . . . . . . . . . . . . 18
3.2 EXTRACTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.3 VACUUM GRADIENT . . . . . . . . . . . . . . . . . . . . . . . . 25
4 CHARACTERIZATION OF ION BEAM AND ION SOURCE . . . . . . . . 27
4.1 PERVEANCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.2 EFFICIENCY OF ION SOURCES . . . . . . . . . . . . . . . . . . 28
4.3 REQUIRED GAS FLOW RATE . . . . . . . . . . . . . . . . . . . 29
4.4 ION FLUX AND NUMBER DESITY OF ION BEAM . . . . . . . 30
5 THE DESIGNED SYSTEM . . . . . . . . . . . . . . . . . . . . . . . . . . 32
5.1 PLASMA GENERATION BY MICROWAVES . . . . . . . . . . . 34
5.2 EXTRACTION UNIT . . . . . . . . . . . . . . . . . . . . . . . . . 43
viii
5.3 MEASUREMENTS OF ION CURRENT . . . . . . . . . . . . . . . 48
6 DISCUSSION AND CONCLUSION . . . . . . . . . . . . . . . . . . . . . 52
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
ix
LIST OF TABLES
TABLES
Table 1.1 Typical Ion Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Table 5.1 Parts and components of the system . . . . . . . . . . . . . . . . . . . . . 33
Table 5.2 Names of the parts in Figure 5.3 . . . . . . . . . . . . . . . . . . . . . . . 34
Table 5.3 Explanation of Figure 5.13 . . . . . . . . . . . . . . . . . . . . . . . . . . 43
x
LIST OF FIGURES
FIGURES
Figure 2.1 Schematic of a DC discharge . . . . . . . . . . . . . . . . . . . . . . . . 7
Figure 2.2 Paschen Curves at the temperature 20oC [8](The graph is a log-log graph) . 10
Figure 2.3 Paschen Curves at the temperature 20oC [9] . . . . . . . . . . . . . . . . . 11
Figure 2.4 Estimated Paschen Curve for H2. (This graph is a semilog graph.) . . . . . 11
Figure 2.5 The T E10 mode of the microwave in a rectengular waveguide . . . . . . . 12
Figure 2.6 Ignition of a microwave plasma . . . . . . . . . . . . . . . . . . . . . . . 13
Figure 2.7 Standing wave pattern within the induced plasma . . . . . . . . . . . . . . 15
Figure 2.8 Location of the plasma tube across the standing wave . . . . . . . . . . . . 16
Figure 3.1 Plasma potential and the regions between the plasma and the wall . . . . . 19
Figure 3.2 A basic extraction system with mesh extractors . . . . . . . . . . . . . . . 22
Figure 3.3 Extraction with the different extraction voltages . . . . . . . . . . . . . . . 22
Figure 3.4 Extraction aperture diagram . . . . . . . . . . . . . . . . . . . . . . . . . 23
Figure 3.5 The extraction diagram for the designed system . . . . . . . . . . . . . . . 24
Figure 3.6 Vacuum zones in order to form the vacuum gradient . . . . . . . . . . . . 25
Figure 5.1 Microwave generator with the integrated power source, the waveguide and
a pyrex plasma chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Figure 5.2 Microwave generator with the integrated power source and the general view
of the assembled system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Figure 5.3 Assembled system without the microwave source . . . . . . . . . . . . . . 34
Figure 5.4 The Pyrex Plasma Chamber; with the gas inlet connector on the left side . 35
Figure 5.5 Plasma generation in oder to test the microwave unit . . . . . . . . . . . . 35
xi
Figure 5.6 Dispersion zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Figure 5.7 The double Langmuir probe . . . . . . . . . . . . . . . . . . . . . . . . . 36
Figure 5.8 The tips of the probe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Figure 5.9 The location of the probe tips . . . . . . . . . . . . . . . . . . . . . . . . 38
Figure 5.10 The Current vs Voltage Graph of 1,5 mbar Microwave Hydrogen Plasma
at the dispersion zone 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Figure 5.11 The Current vs Voltage Graph of 1.5 mbar Microwave Hydrogen Plasma
at the midpoint of the waveguide . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Figure 5.12 The Current vs Voltage Graph of 1.5 mbar Microwave Hydrogen Plasma
at the dispersion zone 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Figure 5.13 Dissembly of the extraction unit . . . . . . . . . . . . . . . . . . . . . . . 43
Figure 5.14 The First Single Aperture . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Figure 5.15 The Conical Single Aperture . . . . . . . . . . . . . . . . . . . . . . . . . 45
Figure 5.16 The Teflon Separator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Figure 5.17 The Vacuum Gradient Part . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Figure 5.18 The Faraday Cup Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Figure 5.19 The Faraday Cup (side view) . . . . . . . . . . . . . . . . . . . . . . . . . 49
Figure 5.20 The Faraday Cup (inside view) . . . . . . . . . . . . . . . . . . . . . . . 50
Figure 5.21 The side view of the simulation of the system . . . . . . . . . . . . . . . . 51
Figure 5.22 The voltage topography of the simulated system . . . . . . . . . . . . . . 51
Figure 6.1 AXCEL-INP simulation for a diode system with three plasma density and
the same voltage drop. From left to right:n1 ≤ n2 ≤ n3 [11] . . . . . . . . . . . . 54
xii
CHAPTER 1
INTRODUCTION
Ion sources started to emerge while Goldstein was working on the canal rays in 1886 before
designing low-current ion sources, which electron-atom collision mechanisms were used for
[1]. In 1930’s, by investigating the arc discharge, higher ion currents started to be provided.
RF and microwave discharges started to be investigated during the next decadeand they were
used in production of ion beams [1]. As it is seen that the origins of the ion sources are
the atomic-nuclear physics research and ion implantation for microelectronic applications;
moreover, ionization sources were developed for space propulsion applications in 1960’s [2].
After all, ion sources have become indispansable parts of particle accelerators and ion im-
plantation systems. For areas of usage, there are many types of ion sources with different
working mechanisms; on the contrary, electron sources have limited variety [1, 2]. The ion
types can be determined for the corresponding applications. Ion sources are mainly used to
produce monoenergetic and unidirectional ion beams [2]. The generated beams are utilized
by guiding them to ion beam lines or by direct guiding them to application zones.
There are plenty types of ion sources. The variety of the ion sources arises from the different
ways of ion generation from solids, liquids and gases and also the variety of generating plasma
such as DC discharge, arc discharge, RF discharge, microwave discharge and laser driven
plasmas [2]. The general types of ion sources are listed in Table 1.1 and these ion sources can
be divided into groups with respect to the ways of ion generation and areas of applications
[1].
1
Table 1.1: Typical Ion Sources
CathodesElectron Bombardment Ion SourcesPlasmatron Ion SourcesMagnetron and Freeman Types Ion SourcesPenning Ion SourcesMulticusp Ion SourcesRF Ion SourcesMicrowave Ion SourcesECR Ion SourcesLaser Ion SourcesElectron Beam Ion Sources /TrapVacuum Arc Ion SourcesLarge Area Ion SourcesIndustrial Ion SourcesLiquid-Metal Ion SourcesPolarized Ion SourcesCluster Ion SourcesIon DiodesIon Sources for Space Applications
The ion sources or ion beam generators have a wide variety of units, such as plasma induction
systems, ion extraction systems, electronic units that induce plasma and supply the extraction
voltage. There are also units for guiding the ion beams. The ion sources necessitate an inter-
acting multi-discipline study which should have the knowledge of plasma physics, electrical
and electronics engineering, and computational systems.
In general, at the end of the ion beam line, there is a linear accelerator. The produced ion
sources are connected to the open-end of this accelerator via vacuum cones. The accelerated
ions are focused by an Einzel lens and dispersed by two quadrupole magnets. Moreover, a
dipole magnet is used in order to deflect the accelerated ions, passing through the ion beam
line. At the other end of the ion beam line, there is a target which is subjected to ion bom-
bardment.
In this thesis, microwave ion sources have been investigated for the design and the construc-
tion. The designed and produced ion sources have been used for the ion beam lines.
In 1977, the first ion source was built by Sakudo and he also developed a microwave ion source
with slit extraction [1]. Ishika, in 1984, used permanent magnets in order to built a compact
2
microwave ion source [1]. Afterwards the gradually increasing interest on microwave ion
sources has spread over, because the microwave ion sources can reach high curret densities at
low pressure plasmas.
The built-in microwave ion source system does not contain any magnetic confinement or any
magnetic support for the plasma, like the electron cyclotron resonance (ECR) microwave
discharges.
The organization of the thesis is as follows: In chapter 2, the theory of ion production via
plasma is discussed and the mechanisms of the direct current (DC) plasma and the microwave
(MW) plasma are reviewed. In chapter 3, the mechanism of the ion extraction is considered.
In addition, the plasma (space) potential theory and the vacuum gradient are discussed. Char-
acterization of ion beams and ion sources are explained in Chapter 4. The main part, the
designed and constructed microwave plasma ion source is expressed in Chapter 5 where the
produced parts of the system are explained. As the last chapter, Chapter 6, the data analysis
of the ion source is done and the theoretical values of the designed system are compared with
the simulation of the designed system and the measured data. Finally results and outcomes of
the thesis are concluded.
3
CHAPTER 2
ION PRODUCTION VIA PLASMA
Altough there are many ways of producing ions, only ion production mechanism via plasma
is the concern of this chapter. Ion production process can be divided into two parts. The first
one is the production of raw ions from a generated plasma and the second is the extraction of
the raw ions from the generated plasma.
2.1 PLASMA
In the universe, more than 99% of observed matter comprises plasma such as interstellar
matter, nebulea, supernova, stars and also the flame [3]. Plasma is the fourth state of matter.
If gases are heated, electrons of the gas molecules start to oscillate and the propability of
ionization of the gas increases. In addition, electrons acquire more energy and their coupled
atoms break up and move around freely. Consequently, the electrons prevail the electrostatic
forces. If a quasineutral gas consists of charged and neutral particles which exhibits collective
behaviour, this quasineutral gas is named plasma [4].
Quasineutrality is defined as approximately equal numbers of negative and positive charges
existing in a system.
The collective behaviour of the plasma can be described by firstly comparing with the ki-
netic theory of gases. The kinetic theory of gases is that no net forces; such as, electromag-
netic forces which act upon the gas particles. Because the gas particles carry no net charges,
they are neutral. So the particles move in straight lines before they collide each other with
a distribution of velocities (Gravitational forces are not our concern at this moment.) [5].
In plasma, charged particles create long range electromagnetic forces. Dominating motions
4
of the charged particles and local charge concentrations affect the whole plasma even if the
local charge concentration is far from the concerning area. While the neutral particles ex-
hibit straight line motions after their collisions with each other, the charged particles exhibit
continuously changing motions because of electromagnetic forces between each other [5].
The electromagnetic forces are long-range forces and interactions which occur between the
charged particles last. These continuous interactions of the charged particles do exhibit non-
linear behaviours. If there is a locally varied charges or a varied current distribution, there
will be impacts on the whole plasma; i.e., the affected charged distribution will evolve into a
new charged distribution.
In addition to the quasineutrality and collective behaviour of the plasma, the temperature, the
number density of the particles and the most importantly, Debye length should be considered
while defining the plasma.
Before defining the Debye length, the Debye shielding should be defined. The Debye shield-
ing is the ability of the plasma to shield out electric potentials that are applied to the plasma
[4]. The electric potentials can be produced by local charges or by inserting electrodes inside
the plasma. The distance, where the potential vanishes, is called the Debye length. The Debye
length, λD, is given by the expressions below [1].
λ2D =
ε0kTe
e2ne, (2.1)
λD = 743
√Te
ne, (2.2)
where k is the Boltzmann constant, ε0 is the vacuum permittivity and e is the elemantary
charge (∼ 1.6 × 10−19C (Coulombs)). The parameters, Te and ne are the electron temperature
in electron volts, the electron number density in inverse cubic centimetre, respectively and the
Debye length, λD is in centimeters [1]. If the Debye length λD increases, the density of the
plasma will decrease; moreover, the Debye length, λD, increases with increasing kTe. The
unit of kTe is Joule.
The Debye length also characterizes the plasma with additional parameters such as, ND, which
is the number of particles in the the Debye region [4]. ND is computed by Equation 2.3 for a
5
Debye sphere which a sphere with a radius of λD.
ND =43
neπλ3D (2.3)
The gas is considered as a plasma when the three conditions, below, are satisfied [4].
λD << L , (2.4)
where L is the size of the plasma system. The dimensions of the plasma region should be
much larger than the Debye length.
ND >> 1 (2.5)
The number of particles should be much more than one in order to consider bulk of the charged
particles as plasma.
ωpτ > 1 (2.6)
Equation 2.5 is the requirement of the collective behaviour. In Equation 2.6, ωp is the fre-
quency of the plasma oscillations and τ is the mean time between the colllisions with neutral
atoms. The plasma frequency should be more than the mean time of the collisions with the
neutrals.
There are various ways to generate a plasma; however, two ways of generating plasma; such
as DC (Direct Current) discharge, and MW (Microwave) discharge, have been mentioned
respectively in this thesis. However, the microwave discharge have been discussed in detail.
2.2 DC DISCHARGE
A DC discharge can be achieved by applying a DC voltage between two conducting electrodes
which are inserted into a gas at low pressures [5]. The typical gas pressure is in between 0.1
torr (0.133322 mbar) and 10 torr (13.3322 mbar) in order to process DC discharge plasma [3].
A schematic drawing of the set-up is given in Figure 2.1.
6
Figure 2.1: Schematic of a DC discharge
Any gas medium contains free ions and free electrons which arise from interactions between
cosmic rays, environmental radiation with the gas atoms or a result of field emission from
any violence on the surface, where electric fields are strong [3, 5]. These free charge carriers
are accelerated through these electric field lines and they start to collide with the atoms or
molecules, encountered with these free charge carriers [3].
These small amount of accelerated free charge carriers loses their kinetic energy by colli-
sions; however, the voltage, applied to the electrodes, maintain these particles to build up
their kinetic energies in order to ionize or excite targets that are the neutral particles, such
as atoms and molecules [5]. If the kinetic energies of the free charge carriers are enough to
exite or ionize the neutral particles, the accelerated charge carriers lose their kinetic energies
by making inelastic collisions with the neutral atoms or molecules; otherwise, the collisions
are elastic if the energy of the free charge carriers have too low to excite or ionize the targets
which are the atoms and the molecules [5]. (Ionization is to break off electrones from the
atoms whereas excitation is to bring electrons of an atom to a higher energy level than the
electrons have.) With the ionization and excitation of the atoms or molecules, the number of
produced charge cariers increases and an avalanche effect occurs. Besides of the free charge
carriers, the produced ions and electrons contribute the excitation or ionization proccess in
the medium. The produced ions and especially the produced electrons gain kinetic energy by
the applied voltage. If there are elastic collisions between electrons and targets, the loss of the
electron energy or the transfered energy can be calculated by Equation 2.7 [5]:
7
Wtr =2me
MTWe , (2.7)
where Wtr is the transfered energy to a target by an electron, MT is the mass of the target,
me is the mass of an electron and We is the energy of the electron. For the inelastic collions
between the energetic electrons and the heavy target, the mechanism and the calculation of
the collision change. The average fraction of the transfered energy is given by Equation 2.8
[5]:
Wtr
We=
MT
me + MT(2.8)
This avalanche effect is caused by the new produced electrons and ions. An electron multiplic-
cation process takes place. The number of electrons N, flowing through the anode electrode
in a unit time, can be calculated by Equation 2.9 [3]
N = N0eαd
1 − γ(eαd − 1), (2.9)
where N0 is the initial number of electrons, α is the probability that the accelerated electron
ionizes a gas atom as it travels a unit distance in the discharge tube (α is the first Townsend
coefficient) [3, 6]. The term eαd is the amplification factor at a distance of d. Beside of
this, it is also the number of ions, produced by the primarily electrons. The coefficient γ is
the efficiency of secondary electrons,e.i., the secondary electrons are produced by the ions,
striking the cathode electrode. It is also called as the second Townsend coefficient [3, 6].
The secondary electron emission coefficient γ is determined by the type of metarial, which
the cathode is made of, and the structure of the cathode surface; in addition to them, type of
gas and reduced electric field E/p have importance [7]. Here, p is the pressure of the region
where the plasma is formed.
The rapid transition, which is from a very poor electrical conductor with the resistivity of
∼ 1014Ωm to a relatively good conductor with a resistivity that is many orders of magnitude
lower, characterizes the breakdown mechanism for the concerning gas, which is in a tube [3].
When the value of N goes to infinity, the electrical breakdown of the gas occurs in the gap
between the two electrodes [3]. The electron amplification factor NN0
, goes to infinity if the
8
denominator goes to zero in Equation 2.9. The condition of the electrical breakdown is given
by Equation 2.10 [3].
γ(eαd − 1) = 1 (2.10)
In order to sustain the DC plasma, the condition, below, should be provided [6].
γeαd = 1 (2.11)
In the case, γeαd → 1 ,the denominator of Equation 2.9 goes to zero and the number of
electrons N goes to infinity. Because of the situation, the electric breakdown occurs.
In order to determine the value of α, α should be written in terms of parameters that change
the value of α. An expression, designated as Equation 2.12, shows dependencies of α:
α = Ape−Bp/E (2.12)
In Equation 2.12, the constants A and B differ for different gases and p is designating pressure
of the gas. The term E determines the electric field of the interelectrode space and therefore
E = V/d [3]. The breakdown voltage or starting voltage Vbr can be computed by combining
Equation 2.12 and Equation 2.10 [3]. In addition to this, the breakdown is sustained at the
room temperature 20oC and the electron mobility is inversely proportional to pressure [7].
Apde−Bp/E = ln (1 + γ−1) (2.13)
By substituting the expression E = V/d into Equation 2.13, the breakdown voltage, Vbr, can
be deduced as
Vbr =Bpd
ln(pd) + ln(A/ ln(1 + 1/γ))(2.14)
9
Expression 2.14, is known as Paschen’s Law [3, 5, 6]. ’pd’ is called as the reduced electrode
distance, on which the breakdown voltage depends. The graph of Vbr versus pd is known as
the Paschen curve. The value of pd plays important role in the extraction process of ions. The
typical Paschen curves of several gases are given in Figure 2.2 [8].
Figure 2.2: Paschen Curves at the temperature 20oC [8](The graph is a log-log graph)
A typical voltage which is needed to sustain the discharge depends upon the type of gas and
the pressure used in the plasma chamber [3]. And as it can be seen in Figure 2.2, the typical
breakdown voltage is at the order of hundreds of volts. However, there is drastic increase at
low pressures. Analysis for the low pressures gives the approximate values. The analysis is
done numerically by using GNU Octave, which is a scientific computing software. The value
of A and the value of ln(A/ ln(1+1/γ)) are taken as 478.680 and 1.664 respectively. In order to
compute the values of A and the value of ln(A/ ln(1+1/γ)), the data points (2.06,1800.00) and
(11.00,259.09) are taken from Figure 2.3 [9] and the estimated low pressure Paschen curve
for the hydrogen gas is drawn as in Figure 2.4 . Note that the unit of pd is in cm-torr in Figure
2.4.
10
Figure 2.3: Paschen Curves at the temperature 20oC [9]
Figure 2.4: Estimated Paschen Curve for H2. (This graph is a semilog graph.)
The breakdown voltage for the low values of pd is very important in order to prevent the
breakdown at the extraction zone while applying high voltages. In addition to these, there is a
crucial concept, plasma potential which should be explained distinctly in order to understand
the ion extraction process.
DC discharge mechanism and DC discharge plasma ion sources are not the concern of this
thesis directly; however, in an ion extraction process, a DC plasma can occur. To prevent the
DC discharge at a vacuum gap, the mechanism of DC discharge should be well understood.
11
2.3 MICROWAVE DISCHARGE
The microwave generated plasmas have similarities with the RF (Radio Frequency) generated
plasmas and optically generated plasmas. The wavelength of the microwaves are at the range
of centimeters, longer than the optical waves which are with wavelengths in nanometer scale
and smaller than the wavelength of RF radiation. Eventually, the all these radiation types are
ordinary electromagnetic radiations, so the plasma generation could be described by the basic
electromagnetic theory.
At first, the electric field of the microwave radiation can be considered in a rectangular waveg-
uide. The microwaves can travel through any closed shaped metals. If these closed shaped
metals are designed for the specific frequency, they guide electromagnetic waves, passing
through them. For the frequency of 2.45 GHz, the rectangular waveguides are used and the
mode of the microwave, which passes through them, is T E10 (Transverse Electric Field Mode
10, in rectancular waveguide). The mode of the wave determines the electric fields and mag-
netic fields of the wave across the width and the height of the waveguide [10]. The electric
field distribution of the mode T E10 is depicted as in Figure 2.5. The electric field has a max-
imum in the center of the width of the waveguide [7]. The other modes can be used any
microwave discharges but they are not the subject of this thesis.
Figure 2.5: The T E10 mode of the microwave in a rectengular waveguide
The free electrons, which are produced by the background UV radiation or gamma radiations,
are made to oscillate by the alternating electric field of the microwave. The T E10 mode is
convenient for plasma generation in the rectangular waveguide becuase at the T E10 mode,
the elecric field reaches its maximum value ath the midpoint of the waveguide [7]. When
the free electrons start oscillating with the oscillating electric field, collisions with the neutral
gas atoms become more violent, so these collisions heat up the gas. Due to the much larger
12
mobility of the electrons with respect to ions , the heating process is accepted to be sustained
by the electrons. The heavier ions can not respond the rapid change in the electric field of the
wave [5]. In this situation, the probability of ionization increases.
Figure 2.6: Ignition of a microwave plasma
The typical electric field strength of a 2.45 GHz microwave is approximately E0 ≈ 30V/cm
[5]. If the collisionless situation( i.e., the electrons do not collide with ions or neutrals) is
considered, the equations of the motion of the electrons can be drived as follows:
E = E0exp(−iωt)x , (2.15)
where E0 is a constant with the same dimensions as E.
ω = 2π f ,
where ω is the angular frequancy of the wave and f is the frquency of a microwave source. In
this work f = 2.45GHz.
mex = −eE , (2.16)
where e is elementary positive charge,
x = −e
meE0exp(−iωt)x (2.17)
by integrating the previous equation, we can get the velocity of the electron,
13
x = −eE0
me
(1−iω
)exp(−iωt)x (2.18)
and by the second integration, we can get the position of the electron,
x =eE0
meω2 exp(−iωt)x (2.19)
The position of the electron is a function of time in Equation 2.19. The position can be
rewritten as follows (Note that the following equation is a scalar function!).
x(t) = x0exp(−iωt) (2.20)
The maximum distance x0 that an electron is driven by the alternating electric field is given
by
x0 =eE0
meω2 (2.21)
for the collisionless situation ν/ω << 1 should be provided where ν is the elastic collision
frequency in gases.
The kinetic energy of the oscillating electron, We, is written as
We =mev2
e
2, (2.22)
where ve is the speed of electron and it is the scalar quantity and the value of ve is the mag-
nitude of x. If the values of a radiation with a frequency of 2.45 GHz and a electric field
strength of 30 V/cm are substituted into the above equations by converting them to SI units,
the maximum distance that the electrons travel x < 10−3cm in a collisionless situation [5].
The maximum enegy that an electron gains in one cycle is about 0.03 eV [5]. This amount of
energy is not enough to ionize the neutral atoms. So if the ignition is to be made at low pres-
sures, there should be some manipulations done to the experiment to increase the probability
of the ionization.
In collisional cases (the cases are at higher pressures than the first situations) it is to transfer
the power from the outside electric field to the unit volume of gas and the average power P is
given by [5]
P =nee2E2
0
2me
( ma
ν2 + ω2
), (2.23)
14
where ma is the mass of the atoms that the electrons collide with. The elastic collision fre-
quency , ν, is approximately in the order of between 109 and 1011 collisions per second at
glow discharge conditions for gases [5].
The ionization of the neutral atoms becomes plausible in collisional situation. Random mo-
tions of electrons stem from the collisions with the atoms of the gas. Moreover, the electrons
gain additional energy from the external electric field during each collision with the atoms.
After an electron makes an elastic collision with an atom, it reverses its direction, at that in-
stant of time, if the electric field changes direction, the electron accumulate enough energy to
ionize the neutral atoms [5].
The plasma is ignited and maintained by the absorption of the electromagnetic energy and the
heat exchange of the thermal plasma is provided by the convective cooling in the gas flow [7].
The incident electromagnetic radiation can only transfer a certain amount of its energy while
travelling through the waveguide. The remaining energy is transmitted forward or reflected
backward to the source by the thermal plasma. In order to increase the dissipated energy by
the thermal plasma, the standing wave is formed by trapping the electromagnetic wave (i.e.,
the microwave.) [7]. As it is shown in Figure 2.7 and also in Figure 2.8, the standing wave
pattern is formed through the waveguide. If special equipments like a stub tunner, a plunger,
an isolater and a directional coupler, are used, the fraction of absorbed electromagnetic energy
can be increased up to 90% to 95% [7].
Figure 2.7: Standing wave pattern within the induced plasma
For the our designed system the location of the pyrex plasma discharge tube, which is almost
15
transparent for the microwaves (instead of the pyrex, quartz is usually used), crosses the
waveguide horizontally.
Figure 2.8: Location of the plasma tube across the standing wave
The microwave has a wavelength of 12.24 cm and the length of the waveguide is three times
the wavelength of the corresponding microwave. There is a standing wave pattern of the
system in Figure 2.8. The source of the microwave is on the left side of the waveguide and
the right side of the waveguide is totally closed and sealed. The pyrex tube is intersecting the
waveguide horizontally between the node and the antinode of the standing wave. At the left
end of Figure 2.8, the oscillation of the electric field is shown. The oscillations of the electric
field in the form of the standing wave induce a plasma inside the pyrex tube as shown.
Before, skipping to a new chapter, there is a very crucial remark that in the absence of mag-
netic confinement for the microwave discharge plasma. There is an upper limit for the electron
number density (i.e., number of electrons per cubic centimeter), ne, in the plasma [1].
ne ≤ 1.11 × 1010 f 2 , (2.24)
where f is the frequency in GHz, and for the microwave with the frequency of 2.45 GHz the
upper limit of the electron density is ne = 6.66× 1010cm−3 [1]. The reason of existance of the
upper limit is that the microwave will not be able to penetrate the plasma without applying
a sufficiently high magnetic field (for f = 2.45 GHz, B = 0.0875 Tesla) through the plasma
when the electron density reaches the upper limit. Moreover, the power of the microwave is
absorbed at the surface of the plasma and the most of it reflects [1].
16
The details of the designed and built system are described in the following chapters with the
help of the theoretical explanations, mentioned above.
17
CHAPTER 3
EXTRACTION OF IONS FROM PLASMA
The produced ions should be extracted from the plasma medium in order to direct them to
targets or guide them to ion beam lines to distort profiles of ion beams. In order to collect
the ions, a positive electrical potential is applied to the plasma medium with respect to the
extracted zone and then ions are ready to be accelerated at the edge of the plasma medium.
Accelerated ions are able to be guided and directed by the electric field and also a dipole
magnet. In order to understand this procedure, plasma (space) potential, extraction should
be discussed firstly. Also, the use of vacuum apertures will be explained for the production
of vacuum gradient.
3.1 PLASMA (SPACE) POTENTIAL
For any practical plasma devices, plasma and walls of the plasma chamber produce a region,
called sheath. Consider a plasma system which has no electric field inside or applied by any
electrodes, inserted in. Since the electrons move much faster than the ions in the plasma, the
electrons leave the plasma at a greater rate than the ions do [4, 7, 11]. The themal velocities
of the individual particles can be calculated for the electrons and ions by using Equation 3.1
and 3.2 respectively.
ve =
√3kTe
me, (3.1)
vi =
√3kTi
Mi, (3.2)
18
where ve is the speed of an electron and vi is the speed of an ion. In Equation 3.2, the
temperature of ions Ti and the mass of an ion Mi should be considered. The value of Mi/me
is about 1837 for a hydrogen ion. From these equations, it can be deduced that the electrons’
thermal velocities are about 1000 times faster than the ions’ thermal velocities [7]. The fast
moving electrons leave the plasma and stick to the walls of the chamber and the slow ions
are assumed to be stationary with respect to the electrons, so that the plasma remains with
a net positive charge [4, 7]. These excess charges produce a potential with respect to the
walls. Debye shielding forms a potential variation in several Debye lengths. This layer is
called plasma sheath [4, 11]. As F.F. Chen mentioned “The function of a sheath is to form a
potential barier so that the more mobile species, usually electrons, is confined electrostatically.
The height of the barrier adjust itself so that the flux of electrons that have enough energy to
go over the barrier to the wall is just equal to the flux of ions reaching the wall [4]”. The
plasma potential drops while approching the wall. An illustration is shown in Figure 3.1.
Figure 3.1: Plasma potential and the regions between the plasma and the wall
The reduction of the potential depends on the configuration and the plasma parameters yet it
is typically 3 or 4 times the electron temperature (in electron volts) per charge of an electron
(in Couloumb) and this situation is also valid for a floating probe or electrode which inserted
in a plasma [11]. The floating electrode or probe can be said to be at the floating potential,
negative with repect to the plasma potential.
19
In a situation that a conducting electrode immersed near the sheath and is held at a lower
potential with respect to the plasma, the electrode attracts the ions and repels the electrons. If
the electron density is taken to be zero at the sheath, the current density J can be written as
[12]. Note that the charge state (or the ionization state) is taken as 1 for convenience while
doing the drivations.
J = nievi , (3.3)
where ni is the number density, e is the elementary charge in Coulombs and vi is the speed of
ions, respectively. The Possion equation is written for this region as
d2Vdx2 =
nieε0
(3.4)
By the conservation of energy (The calculations are non-relativistic calculations!),
Miv2i /2 = eV , (3.5)
where Mi is the individual mass of the ions. The term nie can be written in terms of J and vi,
then the Possion equation becomes
d2Vdx2 =
1ε0
√Mi
2e
√1V
J(x) (3.6)
If Equation 3.6 is multiplied by (dV/dx) and integrated from zero to a potential V [12, 13],
we get
(dVdx
)2
=4ε0
√Mi
2e
√1V
J(x) (3.7)
By taking the square root of both sides of Equation 3.7, and then rearranging it,
dVV1/4 =
2ε0
√Mi
2eJ1/2(x)dx (3.8)
and then by integrating the above equation, the current density is found as
20
J(x) =4ε0
9
√2eMi
V3/2
x2 (3.9)
Equation 3.9 is known as Child-Langmuir Law of space charge-limited current flow [4, 12, 13]
and more general notation for ions having charge state (or ionization state) Z is [2]
J(x) =4ε0
9
√2eZMi
V3/20
x2 (3.10)
The Debye length is the crucial for the boundary of the plasma and for the dynamics of the
transition region. For example, in the case of the high voltage applied to a electrode inside the
plasma, the sheath region gets thicker than the Debye length and the sheath distance dsheath
can be written approximately as [11].
dsheath ∼ λD
√eVkTe
(3.11)
Equation 3.11 becomes very important in order to detemine the probe sizes, dimension of
wire meshes and the extraction apertures and separation [11].
3.2 EXTRACTION
The extraction process of ions from the generated plasma plays a crucial role while developing
an ion source. Ions move randomly inside the plasma with free electrons. Each ion has a
different kinetic energy. The ions should be made mono energetic and unidirectional before
they are injected to a process chamber [2]. This is the reason why multi-farious electrodes
are used. The general types of these are named as accelerating grids or extractor grids or
extractor mesh, (axial) diode system, triode diode extractor, mutigap extraction systems and
multiaperture extractors [1, 11]. All these types are basically simple particle accelerators
with special geometrical design. The extractors should be designed while the limits of the
Chid-Langmuir law is being taken into consideration.
Typical extraction techniques are shown in figures 3.2 , 3.3 , 3.4 [11]. In the present work, the
technique shown in Figure 3.4 is modified. A basic extraction system can be designed as in
Figure 3.2. For this design the target of the ion beams are not defined. For usual purposes, the
21
ions are driven to a vacuum chamber (or downstream region). This vacuum chamber can be
held at a constant electrical potential. For the general case, if the electrical potential is applied
to the vacuum chamber, where the ions
Figure 3.2: A basic extraction system with mesh extractors
impinge the corresponding target, the potential of the downstream region (vacuum chamber)
should be taken into consideration. If the plasma is at the potential, Vpl, and the extractors’
voltages are at the values, Vext1, Vext2, Vext3..., and the potential of the vacuum chamber is at
the potential, Vch, then the general case of a extraction system can be thought as in Figure 3.3
with the given voltages.
Figure 3.3: Extraction with the different extraction voltages
If the charge state Z of the ions are known, the energy of the ions, at the downstream region,
22
can be estimated as
Ei = eZ(Vpl − Vch) , (3.12)
where e is the electron charge in units of Coloumb. Here the plasma potential is increased with
respect to the chamber potential or any electrodes’ potential by applying bias voltage to the
plasma chamber. Nevertheless, the plasma voltage is higher than any potential, applied to the
plasma chamber or any electrodes which are immersed in the plasma. The plasma potential
is 3kTe/e volts or 4kTe/e volts higher than any high voltage electrode or frame, touching the
plasma. It can be a good apprximation when the plasma potential, Vpl, is assumed to be equal
to the potential of a extractor, Vext, which are typically at the order of kilovolts [11]. For low
energy ion sources and their applications, the plasma potential should be added to the voltage
of the extractor and should be taken into consideration [11].
In this chapter, the axial diode extractor system, with a single axial aperture, is going to be
discussed. For this type of extractor, a single aperture is used in order to let the ions go through
it with the diameter, D, and the second aperture is located at a distance, d. Both apertures can
have the same diameter. For the general case, the diameters of the two apertures are different
as they are shown in Figure 3.4.
Figure 3.4: Extraction aperture diagram
For the designed system, the second aperture is narrower than the first single aperture, in
23
order to maintain the vacuum gradient between them. The vacuum gradient is discussed in
the following section. It is worth to mention that the second aperture (at the second grounded
electrode) determines the initial diameter of the ion beam. Plasma potential or the potential
of the electrode, the neighbouring electrode of the plasma, is labeled as PE and the distant
electrode is labeled as GND which means the grounded electrode in Figure 3.4.
This diode extractor is placed in front of the plasma source. It is important to be near the
plasma. This extraction system is located as shown in Figure 3.5. The vacuum chamber is
located after the diode extractor. A high voltage (about 3 kV) is applied to the first aperture
electrode and the second aperture electrode and the vacuum chamber are grounded as they are
shown in Figure 3.5. The potential of the plasma, Vpl, is assumed approximately equal to the
potential of the first extractor, Vext1 (Vpl = Vext1). Both the potential of the second extractor,
Vext2, and the potential of the vacuum chamber, Vch, are grounded. Then the system becomes
as it is shown in Figure 3.5. The energy of the beam ions is given by,
Ei = eZVext1 (3.13)
Figure 3.5: The extraction diagram for the designed system
Figure 3.5 is the schematic diagram of the designed system. For the rest of the thesis, discus-
24
sions would be based on this configuration. The importance of ion extraction will be discussed
in Chapter 4.
3.3 VACUUM GRADIENT
Vacuum process is very impotant in this system. Unless the plasma is generated at a specific
pressure, it will not be sustained and will not be stable. Moreover, the extraction process of the
ions should be succeeded at very low presures in order to prevent the DC dischage between
the extraction electrodes and the metal frames of the system.
In the designed system, the plasma chamber (tube) is held at a pressure of Pplasma. This
pressure defines the plasma pressure. The place where the extraction electrodes are seperated,
is held at a pressure of Pextraction which is lower than the plasma pressure, Pplasma. Finally,
the ion beam line or the measurement vacuum gradient part ends with a place which is held
at the lowest pressure, named as Pbeamline. The scheme is shown in Figure 3.6. The plasma
pressure is controlled by the gas inlet which is located at the left side of the figure. The gap
between the extraction electrodes is vacuumed by the rough vacuum. The rough vacuum of
the whole system is also made from this connection as it is shown in Figure 3.6. The system
is brought to the highest vacuum levels by the high vacuum pumps throught the turbo vacuum
connection.
Figure 3.6: Vacuum zones in order to form the vacuum gradient
25
Finally, by applying the vacuuming process part by part, the vacuum gradient is sustained.
To maintain the vacuum gradient, the separation of the apertures and the exraction aperture
should be small enough and also the geometry of the apertures is very important. The conical
separator is used to achieve the vacuum gradient in the system. In addition to these, the
number of the vacuum separators between the vacuumed parts is also significant.
26
CHAPTER 4
CHARACTERIZATION OF ION BEAM AND ION SOURCE
4.1 PERVEANCE
Perveance defines the relation between the electron or ion beam current and the extraction
voltage of the source with fixed cross-sectional beam area [2]. The perveance is given as
P ≡Ib
V3/20
=JπD2
4V3/20
(4.1)
In the above equation, Ib is the beam current and V0 is the exraction voltage. D is defined as
the initial beam diameter whereas J is the density of the beam current (look at Figure 3.4).
For ideal ion sources, the extraction voltage would be applied via planar accelerating grids
seperated by a distance d. The extraction voltage V0 would operate with the accelareting
grid with no webbing [2]. The ion current density can be written via the Child space-charge
limited condition.
J =4ε0
9
√2eZMi
V3/20
d2 (4.2)
Here, Z indicates the charge state (or ionization state) of the ions. The Child-Langmuir law
perfectly suits for estimating the maximum value of ion beam current [11]. More practical
form of Equation 4.2 is given by [1]
J = (1.72)
√Zu
V3/2
d2 (4.3)
27
where J is in units of miliamper per square centimeter (mA/cm2), u is the atomic mass unit of
the ions, V is in units of kilovolt (kV), and d is in the unit of milimeter (mm).
The ion beam current Ib, is determined by the smallest aperture of the extraction electrode
and by multipying the current density J, with the area of the aperture, πD2/4. The net current
is found as
Ib =πD2
4J (4.4)
Ib =ε0π
9
√2eZMi
V3/20 D2
d2 (4.5)
For an ideal ion source the perveance is found by substituting Equation 4.2 into Equation 4.1
as,
Pmax =JπD2
4V3/20
=πε0
9
√2eZMi
(Dd
)2(4.6)
It is better to design an ion source, having the value of the perveance which should be close
the theoretical value of the perveance Pmax. In this work, the theoretical value of Pmax is
calculated as 2.53 × 10−10(A/V3/2) with respect to the dimensions of the designed extraction
unit.
4.2 EFFICIENCY OF ION SOURCES
The power consumpsion of an ion source Pis, can be written by the sum of the power con-
sumption of the extraction unit Peu, and the power consumption of the plasma generator Ppg.
Pis = Peu + Ppg (4.7)
the power consumption of the extraction unit is given by [2],
Peu = IbV0 (4.8)
28
So Equation 4.7 becomes
Pis = Ib(V0 + V ′) , (4.9)
where V ′ is the required potential (V) by the source to produce the ions and Ib is the ion
current [2]. V ′ can be obtained by measuring the power consumption of an ion production
unit which is used to produce raw ions for the system in order to obtain the ion current Ib.
The power consumption of the ion production unit is calculated by Wipu = IbV ′ for electrical
devives. Where Wipu is the dissipated power by the ion production unit for the corresponding
ion current Ib. The electrical efficiency of the ion source, ηe, can be written as
ηe =Peu
Pis=
V0
V0 + V ′(4.10)
or alternatively,
ηe =1
1 + (V ′/V0)(4.11)
4.3 REQUIRED GAS FLOW RATE
The imperative ingredient to provide the required plasma density is the gas . Firstly, the ratio
of the ions’ out flow from the source to the in-flow of the neutrals into the source determines
the the gas utilization efficiency, ηg [2].
ηg =S i
S n=
Ib
eS n, (4.12)
where S i and S n are the out-flow and the in-flow rates, respectively. The term, S n, is the flow
rate of the neutrals into the plasma chamber.
The net efficiency of the ion source ηis is calculated by multiplying the gas utilization effi-
ciency ηg, with the electrical efficiency ηe [2].
ηis = ηgηe (4.13)
29
If the in-flow rate of the gas is too low, it causes an extra load to the plasma generator and the
efficiency decreases. The good utilization of the gas, for supplying the propellant to the ion
source, is a very important effect on the ion sources.
4.4 ION FLUX AND NUMBER DESITY OF ION BEAM
In order to find out the number of the particles passing through a region per unit time, the ion
flux should be considered. The ion flux, φi, of the system can be written by using the initial
beam diameter, D [2].
φi =J
eZ(4.14)
and
J =4Ib
πD2 (4.15)
by substituting Equation 4.15 into Equation 4.14
φi =4Ib
πeZD2 (4.16)
φi is in units of ions per square meter per second (ions/m2s). Moreover, the number density of
the ion beam is calculated by using Equation 4.4 [2], and rewriting the equation for nb which
is the number denisty of the ion beam:
J = Zenbvb , (4.17)
nb =J
Zevb(4.18)
or
nb =φi
vb, (4.19)
where vb is the mean velocity of the ions for the applied extraction voltage, V0. From the
conservation of energy (here the thermal energy of the ions are neglected) the mean velocity,
vb, is written as
30
vb =
√Mi
2eZV0(4.20)
If equations 4.2 and 4.20 are substituted into Equation 4.18, the number denisty of the ion
beam, nb, is found as
nb =4ε0V0
9eZd2 (4.21)
As it can be seen explicitly in Equation 4.21, the ion density is independent of the ion mass
and the unit of the number density is in ions per square meter (ions/m2).
31
CHAPTER 5
THE DESIGNED SYSTEM
A microwave coupled plasma ion source is designed and constructed in this work. The mi-
crowave generation is provided by a magnetron which has the output power of 700 Watt with
the frequency of 2.45 GHz. Since the high voltage power supply of the magnetron alternates
with the frequency of 50 Hz, the generated microwaves pulse with the frequency of 50 Hz.
The power supply of the magnetron has a half-wave rectifier, so the magnetron also turns off
after each cycle of the alternating current. When the magnetron is turned on, the microwave
generation starts with the maximum power. The magnetron is cooled by air ventilation. The
3D design of the microwave generator is shown in Figure 5.1.
Figure 5.1: Microwave generator with the integrated power source, the waveguide and a pyrexplasma chamber
The dimensions of the waveguide are determined with respect to electromagnetic waves with
the frequency of 2.45GHz which sits in the microwave region of the electromagnetic spec-
32
trum. The inside height of waveguide is 45 mm whereas the inside width is 93 mm. Because
of the type of the rectangular waveguide, the mode of the propagating wave is T E10 inside the
waveguide.
As they are shown in Figure 5.2, all the parts of the system numbered and titles of the num-
bered parts are given in Table 5.1.
Figure 5.2: Microwave generator with the integrated power source and the general view ofthe assembled system
Table 5.1: Parts and components of the system
1 Gas Inlet2 Plasma Chamber (Pyrex Tube)3 Waveguide4 Extraction Electrodes5 High Voltage Cables6 Vacuum Gradient7 Faraday Cup
In Figure 5.3, the system is unmounted from the waveguide and the parts are named as inTable 5.2.
33
Figure 5.3: Assembled system without the microwave source
Table 5.2: Names of the parts in Figure 5.3
1 Gas Inlet2 Plasma Chamber (Pyrex Tube)3 Extraction Unit4 Vacuum Gradient5 Faraday Cup Unit6 Vacuum Line (for the extraction unit)7 Vacuum Line (for vacuum gradient)
5.1 PLASMA GENERATION BY MICROWAVES
The microwave ion generation unit of the system comprises of a waveguide with the length
of 375 mm and a pyrex plasma chamber, mounted through the waveguide horizontally. The
circular holes that hold the plasma tube have a diameter of 30 mm in order to mount the pyrex
plasma chamber tightly.
34
Figure 5.4: The Pyrex Plasma Chamber; with the gas inlet connector on the left side
The ends of the waveguide are closed and sealed in order not only to produce standing wave
pattern inside the waveguide, but also to prevent microwave leakage. The generated mi-
crowaves are pumped in near the end of the waveguide via the microwave antenna of the
magnetron. The length of the waveguide is approximately 3 times longer than the wavelength
of 2.45 GHz microwave. The plasma is generated between the nodes of the standing waves.
In Figure 5.5, the photograph of the microwave induced plasma system is shown while the
microwave induced plasma is forming through the pyrex plasma tube.
Figure 5.5: Plasma generation in oder to test the microwave unit
The plasma characterization is done by using a double Langmuir probe. The data of the
current is measured by Keithley-2400 Sourcemeter and is collected by Labview-9 via GPIB
(General Purpose Interface Bus) interface. The double Langmuir probe is inserted into the
plasma at a horizontal position through the cylinderical symmetry axis of the pyrex chamber
35
in order to find out the plasma parameters, such as the electron density, ne, and the electron
temperature, Te. The data sets are taken at the dispersion zone of the plasma. The dispersion
zone is shown in Figure 5.6, where the plasma disperses to the ends of the chamber when it
pervades from borders where the waveguide encompasses the chamber.
Figure 5.6: Dispersion zone
In order to measure the plasma parameters, the double Langmuir probe, used in this system,
is shown in Figure 5.7. The double Langmuir probe was prepared especially for this system.
The probe has a pyrex probe tube, prepared as the body (1) of the probe. The ends of the
body are sealed. Two copper wire lines (2) cross through the pyrex body and are connected
the tungten wire lines which end at the tips of the probe. The connection part (3) is made
of aluminium and it connects the probe to the plasma chamber (tube). There is a O-ring,
got clamped inside the connection part. This clamped O-ring allows to adjust the location
of the probe tips in the plasma chamber by moving through the axis of the body (1) and also
prevents gas leakage inbetween the body(1) and the connection part (3). The tips (4) are made
of tungsten.
Figure 5.7: The double Langmuir probe
36
There is a closer view of the double Langmuir probe in Figure 5.8. The two tips are separated
from each other by two pyrex tube branches and the tungten wires are soldered to the copper
wires at the joining point of the two branches. The length of these branches are 6.5 cm long
approximately. The diameter of the each probe tip is 0.25 mm and the length of the tips is 10
mm. The total surface area of the each pobe is 7.9mm2.
After the tips of the probe are immersed into the plasma, a voltage source is connected to the
electrodes of the tips. The voltage is increased gradually from -70V to +70V. The current and
the voltage difference between the two tips are measured and saved at the computer.
Figure 5.8: The tips of the probe
The tips of the probe are localized at 50 mm away from the center of the waveguide’s sym-
metry axis. The localization of the tips of the probe are shown in Figure 5.9(a) and in Figure
5.9(b). In Figure 5.9(c), the double Langmuir probe is mounted and operated and it can been
seen as a silver color cylinderical metal on the right-hand side of the plasma chamber. In
order to prevent the noise stemming from the microwave with the frequency of 2.45 GHz, the
tips of the probe are located out side of the waveguide. If the data are taken at the position
mentioned above, the comparison can be made consistently while the plasma parameters are
investigated at the midpoint of the waveguide. Moreover, a filtration process is applied while
the plasma is forming. Low pass filter is used at the frequency of 0.1 Hz.
The values of the electron number density ne and the electron temperature, Te at the midpoint
of the waveguide are expected to be more than the values, measured at the dispersion zone,
which is the outside of the waveguide.
37
(a) The plasma is off (b) The plasma is on
(c) whole view
Figure 5.9: The location of the probe tips
The current versus voltage graph of the microwave induced plasma, at the pressure of 1.5 mbar
(1.13 torr), is shown in Figure 5.10. The plasma electron temperature, Te, and the electron
denisty, ne, are calculated by using the following equations [14]:
dIprob
dVprob
∣∣∣∣∣Iprob=0
=12
eI+
kTe, (5.1)
Te =12
ek
∆Vd , (5.2)
where A is the area of the individual tip surface, Vprob is the voltage difference between the tips
of the double probe, Iprob is the probe current, I+ is the saturation current, k is the Boltzmann
constant, and m+ is the mass of the hydrogen ion H+.
38
Figure 5.10: The Current vs Voltage Graph of 1,5 mbar Microwave Hydrogen Plasma at thedispersion zone 1
By using the graph, above, the probe current Iprob is obtained as ∼ 250µA and the ion sat-
uration current, I+, is determined as ∼ 200µA. In order to determine the slope of the linear
part of the curve at Iprob = 0, the value of the probe current, at 250µA, and the value of
the probe voltage, Vprob, at 4.38V can be used. By substituting the values of the probe cur-
rent Iprob =∼ 250µA and the corresponding voltage, Vprob = 4.38V , into Equation 5.1, the
expression can be written as
250 × 10−6
4.38=
12
ekTe× 200 × 10−6 (5.3)
If the above equation is arranged for Te, then Te is found approximately ∼ 2eV (∼ 23200oK).
ne =I+
0.61eA√
kTe/m+
(5.4)
The ion saturation current, I+, can be determined by drawing a tangential line horizontally to
the first crest of the ion current decline at the positive voltage region. This is done for fitting
and determining the tangent hyperbolic characteristic of the double probe’s I vs V curve. It
should not be forgotten that this procedure is approximate determination of the ion saturation
39
current. If the ion saturation current, I+, is ∼ 200µA is substituted into Equation 5.4 [14],
Equation 5.4 becomes
ne =200 × 10−6
0.61 × 1.6 × 10−19 × 7.9 × 10−6√
1.38 × 10−23 × 23200/1.6 × 10−27(5.5)
Finally, the electron density, ne, is calculated for the pressure of 1.5 mbar to be 2 × 1016m−3
or 2 × 1010cm−3.
By repeating the experiment, the following data were taken and the graph is shown in Figure
5.11, in order to obtain the plasma electron temperature and density at the midpoint of the
waveguide. The pressure was held at 1.5 mbar and again the hydrogen gas was used. By
using the graph, in Figure 5.11, and by doing the same calculations, mentioned previously,
the electron temperature, Te, is obtained as ∼ 8eV (∼ 92800oK), and the electron denisty, ne,
is obtained in the order of ∼ 7.5 × 108cm−3. As it can be seen from the value of the electron
number density, ne, it is smaller than the value which is found for the data of the dispersion
zone, is in the order of 1010cm−3.
Figure 5.11: The Current vs Voltage Graph of 1.5 mbar Microwave Hydrogen Plasma at themidpoint of the waveguide
As it is seen in Figure 5.11, the I vs V curve of the probes is not symmetric about the origin.
40
And also the found ion density is smaller then the dispersion zone 1. It is not an expected
value for the core of the plasma.
The experiment was repeated again in order to obtain the symmetric I vs V curve of the plasma
outside of the waveguide. The probe tips were at a distant of 57 mm from the midpoint of
the waveguide. The reason of this is to observe the electron density decrease while moving
the probe to the outside of the waveguide. The data were taken for the microwave hydrogen
plasma at the pressure of 1.5 mbar. The graph of this process is shown in Figure 5.12.
Figure 5.12: The Current vs Voltage Graph of 1.5 mbar Microwave Hydrogen Plasma at thedispersion zone 2
If the data analysis above is applied, the electron temperature, Te, is obtained as ∼ 5eV
(∼ 58000oK), and the electron denisty, ne, is obtained in the order of ∼ 109cm−3 at the distant,
57mm far from the midpoint of the waveguide and in the vicinity of the extraction electrodes.
As it can be seen from the result of the calculation, the electron density of the plasma decreases
when the plasma disperses. However, the second I vs V Graph in Figure 5.11 is not as accurate
as the first I vs V Graph, in Figure 5.10. The reason of this is that the second I vs V Graph is
not symmetric with recpect to the origin and there is a shift on the graph. While the data are
been taken out side of the waveguide, at the dispersion zone 2, the plasma can be affected by
41
the environmental impacts. And also at a little far distant, out of the waveguide, the plasma
becomes more instable. Since the microwaves can not provide the sufficient energy to the
molucules in order to ionize them at that region. The plasma starts flickering far from the
waveguides borders. The waved shapes, at the left and right sides of the curve, can be caused
by these instabilies at the dispersion zone 2. The region, where the data are taken accurately
is very narrow in this experiment. However, there is some noise, observed in the waveguide
that is not eliminated. Consequently, for the data of the dispersion zones, the current readings
and the voltage readings are more accurate than the data of the midpoint zone. While taking
the data sets of the experiment, the noise factors are discarded by low pass filters via the
computer. The filter tools are used to get rid of the noise factors in the computer programme.
However, these filtering tools are not electronic tools, they are only the computational tools
that reduce some of the unwanted signal while the I vs V Graphes are being drawn and they
are not enough to get rid of the noise and the other effects which occured at the midpoint of
the waveguide. The other effects are discussed at the conclusion part of the thesis. While
doing the experiment, the data, which were obtained for the 1.5 mbar microwave plasma at
the dispersion zones, have been used in the calculations and comments.
42
5.2 EXTRACTION UNIT
In this experiment, the diode extraction system is used with the single aperture. There is a
disassembled view of the extraction unit in Figure 5.13 and the parts are named as in Table
5.3.
Figure 5.13: Dissembly of the extraction unit
Table 5.3: Explanation of Figure 5.13
1 Plasma Chamber Connection2 First Single Aperture3 Teflon Separator4 Conical Single Aperture5 Vacuum Gradient Connection6 Vacuum Line
The plasma chamber connection is made of aluminium in the shape of a disk which joins the
43
first single aperture with four bolts and an O-ring is placed inbetween the plasma chamber
connection and the first single aperture. All the o-rings are lubricated with low pressure
lubricant.
In Figure 5.14(a), the plasma side of the first single aperture is shown. This side faces the
plasma chamber and the plasma chamber is mounted in this side. In Figure 5.14(b), the teflon
separator side is shown. This side faces the teflon separator and presses the teflon separator
with four long bolts which connect the first single aperture, the teflon separator and the conical
single aperture together. The side, facing the teflon separator, is smooth and flat to provide a
proper vacuum. The diameter of the aperture is 4 mm and the thickness of the aperture has a
width of 1 mm, approximately.
(a) Plasma Side (b) Teflon Separator Side
Figure 5.14: The First Single Aperture
The conical single aperture is shown in Figure 5.15. The conical side faces the teflon separator
and it has an aperture, with the diameter of 2 mm, at the tip. The reason of the conical shape
is to produce curved electric field lines in order to guide the accelerated ions to the aperture.
In Figure 5.15(b), the teflon separator side of the conical aperture is shown. This side has a
smooth and flat surface, like the teflon separator side of the first single aperture, in order to
make perfect contact with the teflon separator. The other side of the conical single aperture
has a cavity (in Figure 5.15(c)), like the first single aperture has also, in order to join the
vacuum gradient part. And this side is mounted to the vacuum gradient part with an O-ring
and a aluminium connection disk. Both the first single aperture and the conical single aperture
are used as the electrodes in order to apply the extraction voltage.
In Figure 5.16, one side of the teflon separator is shown. The teflon separator is used for an
isolator in between the two extraction electrodes, which are also the apertures of the exraction
unit. Besides of being used as an isolator, it is very durable for the vacuum and is convenient to
44
be processed and shaped. The teflon separator has O-rings and cavities on its both sides. The
O-rings are placed in the coves and they make perfect connections with the first single aperture
and the conical single aperture. The teflon separator also has a vacuum output individually at
the bottom of it. The vacuum output is also shown as a red vacuum pipe in Figure 5.16. The
blue plastic pipes are used to isolate the bolts and nuts in order to prevent any short circuits
between the first single aperture and the conical single aperture.
(a) Profile (b) Teflon Separator Side
(c) Vacuum Gradient Side
Figure 5.15: The Conical Single Aperture
Figure 5.16: The Teflon Separator
The vacuum gradient part is shown in Figure 5.17 with its connection disks and a vacuum
pump connector. The left side of the vacuum gradient part is connected to the conical single
45
aperture with an O-ring and a connection disk. The right side of it is connected to the Faraday
cup in order to measure the ion beams. In addition to them, the connection of the underside
joins with the high vacuum pump in order to produce very low pressure gradient. While the
pressure, inside the teflon separator, is in between 10−1 and 10−2 mbars, the inside of the
vacuum gradient part can be at the pressure of the order of 10−3 mbars.
Before starting the experiment, the whole system is cleaned in order to use hydrogen gas in
the experiment. After that, the whole system pressure is brought to a desired pressure and
then the vacuum gradient side is brought to lower pressures to elongate the mean free path
of the ions. So they can travel more distance without collisions. If the inside of the teflon
separator is held at a constant pressure, the vacuum gradient will be sustained through the
whole system, from the gas inlet to the target.
Figure 5.17: The Vacuum Gradient Part
The vacuum gradient part plays a very crucial role in this experiment. If the pressure is not
low enough inside the vacuum gradient part, which takes place in between the conical sin-
gle aperture and the Faraday cup, a DC discharge can be induced between the conical single
aperture electrode and the metal flange of the Faraday cup. It is important to prevent this
46
DC discharge for measuring the accurate data of ion current, produced by the microwave in-
duced plasma and its corresponding extraction unit. The pressure, inbetween the extraction
electrodes, changes from the order of 10−1 mbars to the order of 10−2 mbars (from the order
of 10−2 torr to the order of 10−3 torr). The estimated Paschen curve of the hydrogen gas, in
Figure 2.4, is used for determining the critical value of the DC breakdown. If the highest
pressure value, 10−2 torr, is taken into account, the value of pd becomes 5 × 10−3 torr-cm
for the minimum separation of 5 mm, between the extraction electrodes. If the corresponding
breakdown voltage is found from Figure 2.4, it is seen that the value of the breakdown voltage
exceeds the value of 10kV. In the designed system, merely 3kV is applied between the extrac-
tion electrode and the Faraday cup. Therefore, the occurance of the DC discharge plasma is
prevented.
47
5.3 MEASUREMENTS OF ION CURRENT
While measuring the ion current, a Faraday cup is used. A Faraday cup is a device which
measures the current of charged particles. The basic theory of the Faraday cup is that a cup
shaped metal faces the incoming charged particles. The impinging charged particles charge
up the Faraday cup and it gains a net charge. The Faraday cup discharges this small amount
of the ion accumulation by grounding. If this small amount of current is measured by an
electrometer then the current of the ions, entering the Faraday cup, is determined. The number
of ions per second, entering the Faraday cup, N is given by,
N = It/e , (5.6)
where I is the measured current (in Amperes), t is the time interval(in seconds) when the
measurement is done, and e is the electronic charge (in Coloumbs). The above equation can
be assumed as a good approximation for the ions, having the ionization state of one.
Figure 5.18: The Faraday Cup Scheme
If a resistor is connected between the Faraday cup and the ground in series, the voltage drop
can be measured over this resistor and the amount of current can be calculated in order to
neutralize the incoming beam, impinging into the Faraday cup. An electrometer can also be
connected in series instead of the resistor. The Diagram is depicted in Figure 5.18. Figure
5.18 also shows the principal circuit diagram of the measurement unit of the microwave ion
source. In this system, there is a resistor with a resistance of 100 kΩ in series and the voltage
is read on this resistor by a scope.
48
For the measurements of this microwave ion source, the Kimball Physics model FC-73A
Faraday cup is used. The Faraday cup consists of two parts as it is shown in figures 5.19
and 5.20. In Figure 5.19, the part, numbered as (1), is the converter and connection part of
the Faraday cup, joining the vacuum gradient part. And the second part, numbered as (2),
is the head of the Faraday cup. At the most-right of the Faraday cup, there are three BNC
connections which are used for different purposes. The current reading BNC connection of
the Faraday cup is numbered as (3). Both the converter-connection part and the head of the
Faraday cup is made of stainless steel. A copper O-ring is placed between the part (1) and the
part (2) while joining them together.
Figure 5.19: The Faraday Cup (side view)
The profiles of the parts are shown in Figure 5.20. The head of the Faraday cup has an aperture
with the diameter of 5 mm. The incoming ion beam impinges on this aperture. The area of
this aperture is 0.196cm2.
49
Figure 5.20: The Faraday Cup (inside view)
In this work, the microwave induced hydrogen plasma was generated at the pressure of 1.5
mbars and the extraction voltage, Vext was applied to the first single aperture at a voltage of
3kV . Moreover, the conical single aperture and the frame of the Faraday cup were grounded.
The vacuum gradient part was kept at a pressure of order of 10−3mbars. At this plasma pres-
sure and extraction voltage, Vext, the impinging ion current was measured as 25µA, passing
through the aperture of the Faraday cup with an area of 0.196cm2. The total current, passing
through per centimeter square was calculated as 125µA.
The trajectory of the generated ions were simulated with the actual dimensions of the system.
The simulations were done with the computer programme, known as SIMION Version 8.0.
The simulation of the ion trajectories were made for 20 protons, in the vicinity of the first
single aperture without any thermal energies. The protons were accelerated by 3kV . The
aperture of the Faraday cup was located at the most-right end of the simulation screen. For
low densities, the focusing of the particles can be achieved by the geometry of the extraction
apertures and the extraction electrodes. The side view of the simulation of the system is shown
in Figure 5.21.
50
Figure 5.21: The side view of the simulation of the system
In Figure 5.22, the voltage topography of the simulation can be seen; in addition to this, the
motion of the particles can be easly perceived through the voltage downstream.
Figure 5.22: The voltage topography of the simulated system
51
CHAPTER 6
DISCUSSION AND CONCLUSION
In this thesis, the microwave induced plasma ion source was investigated in detail. In order to
achieve this, the ion production via plasma by using a microwave radiation and ion extraction
systems were discussed. Moreover, microwave source design and plasma induction systems
were studied. In order to obtain and to have the knowledge of the plasma parameters, the
double Langmuir probe techniques were considered and applied. The vacuum gradient was
constituted and used in order to increase the mean free path of the extracted ions.
In order to prevent the electrical breakdown between the extraction electrodes the estimated
Paschen curve for the hydrogen gas, in Figure 2.4, was drawn and used to verify the clear
extraction without any breakdown. The estimated value of the breakdown voltage is more
than 10 kV from the estimated graph in Figure 2.4, for the minimum separation of 5 mm and
it is also consistent with Equation 6.1 [1].
d ≥ 1.41 × 10−2V (6.1)
where d is the separation distance of the extraction electrodes, in units of milimeter (mm) and
V is the extraction voltage, in units of kilovolt (kV) between the electrodes. For this designed
system, the relation was found for the minimum separation gap, d = 5mm, and the extraction
voltage, V = 3kV , as
5mm ≥ 1.41 × 10−2(3kV)
5mm ≥ 7.3 × 10−2mm
52
Our designed and constructed system worked as it was expected. However, some errors oc-
cured while taking the data at the midpoint of the waveguide. Before discussing the main
errors, the known errors should be made clear.
There was an anomaly for the plasma density at the midpoint of the waveguide. The electron
density was decreasing, when the probe tips were drawn out of the waveguide as it was ex-
pected. However, the data measurements at the midpoint were not accurate. The calculated
value of the density is in the order of 108cm−3 and it is much smaller than the values, which
were calculated for the dispersion zones. There is a contradiction and, to analyse this situa-
tion, a good filtration and an isolation should be done for the double Langmuir probe. When
the probe tips were immersed into the waveguide area, the microwave radiation could cause
sparks or generate plasma regions in between the pyrex probe tube through which the tungsten
probe wires crossing. In addition to these, there could be one more reason. The density of the
particles decreases in a great deal while they spread towards the waveguide borders. And also
the TEM mode of the microwave sustains this distrubution while it decreases its field strength.
Moreover, the one of the nodes of the standing wave was close to the pyrex plasma tube, the
tips were not located accurately. If the tips had been moved a little bit out of its determined
location, the data would not been measured accurately. The dislocating the tips is the other
reason of not measuring the maximum values of the plasma parameters at the midpoint of the
waveguide where the plasma parameters reach the maximum values as they are expected to
be. When the probe tips were drawn away from the waveguide borders, where the dispersion
began, the electron density decreased as it was expected.
As a result, the microwave ion source system was designed and constructed by generating
the hydrogen plasma. The plasma was generated by using the microwave source with the
frequency of 2.45 GHz and the power of 700 Watt. The generated microwave was pulsing
with the frequency of 50 Hz. The plasma parameters were taken by the double Langmuir
probe and the ion current was measured by the Faraday cup. The designed and constructed
microwave system generated the hydrogen plasma with ∼ 1010cm−3 electron density (ne) and
∼ 2eV electron temperature (Te) in the pyrex chamber (tube) at the pressure of 1.5 mbar in
the region, out of the microwave-guide, where the plasma dispersed.
By refering Equation 6.2, there is an upper limit for the electron density, ne, of the plasma
and it is about ne = 6.66× 1010cm−3. For the calulated values from the I-V curves, mentioned
53
above and also in ’Chapter 5’, the maximum value of the electron density was found as 2 ×
1010cm−3 at the outside borders of the waveguide, so the density values verify the theoretical
inequality, mentioned below. And the density also decreases rapidly in few milimeters from
1010cm−3 to 109cm−3 order.
ne(= ni) ≤ 1.11 × 1010 f 2 (6.2)
2 × 1010cm−3 ≤ 6.66 × 1010cm−3
For the ion extraction part, an ion current was generated by the designed system and the
measurements were done by the Faraday cup. The ion current was measured as 125 µA/cm2
at a distance of approximately 125 mm away from the second extraction electrode by applying
the extraction voltage of 3 kV to the first single aperture electrode and grounding the conical
single aperture and the frame of the Faraday cup.
In Figure 6.1 [11], three ion extraction processes are given with different plasma densities.
The densities increase as n1 ≤ n2 ≤ n3 and the extraction voltage is held constant. If the
plasma density is too low then the plasma border (plasma meniscus) forms a concave shape
with the extraction voltage. With increasing density this concave shape turns out to be a
convex shape. After the second electrode, the ions travel a distance, they repel each other and
there occurs a divergence because the electric field is created by the ions. As a result, the ions
repels each other due to this electric field.
Figure 6.1: AXCEL-INP simulation for a diode system with three plasma density and thesame voltage drop. From left to right:n1 ≤ n2 ≤ n3 [11]
In figures 5.21 and 5.22, the convex shape is seen clearly before they leave the second elec-
trode and the divergence is formed after the second electrode. This divergence can be con-
trolled by a triode extraction system and by an Einzel lens. If a collimation is done by this
way, the beam quality can be increased. The Einzel lens system should be added to the beam
line to provide the collimation through the beam line.
54
For the designed system, the parameters of the ions, such as the perveance, P, so the ion cur-
rent, Ii should be compared with the value of the maximum perveance, Pmax, so the maximum
extractable ion current, Iimax. The values of the parameters are calculated by
Pmax =πε0
9
√2(1.6 × 10−19C)(1.67 × 10−27kg)
(1mm13mm
)2
(6.3)
= 2.53 × 10−10(A/V3/2)
If the maximum perveance, Pmax, is multiplied by the extraction voltage, V0, of 3000 volts,
the maximaum ion current, in the vicinity of the conical single aperture, is found:
Iimax = 2.53 × 10−10 × 30003/2 (6.4)
= 4.16 × 10−5(A) = 41.6(µA)
The measured value of the ion current is 25µA. If the theoretical and measured values are
compared, the ion efficiency, ηa−a, between the conical single aperture and the aperture of the
Faraday cup is found as:
ηa−a =25
41.6(6.5)
= 0.60
The percentage of the ion efficiency is 60%. Also it can be seen from Figure 5.21 that 12
of 20 particles can reach into the middle of the grounded electrode where the measurements
were taken. The measured value, above, is consistent with the value, which is found from
Figure 5.21. These values, for the ion current readings, were done without applying any
Einzel lensing or any collimation process for the extracted ion beam.
For the further research, the relations between the microwave plasma parameters and the
perveance parameter can be invesitaged by changing the extraction voltage, plasma pressure.
In addition to them, the design of the extraction apertures can be changed and effects on the
perveance can be investigated further.
55
REFERENCES
[1] B.H. Wolf. Handbook of Ion Sources. CRC Press, Inc., Boca Raton ;New York; London;Tokyo, 1995.
[2] J.R. Roth. Industrial Plasma Engineering, volume 1. Institute Of Physics Publishing,Bristol, 2001.
[3] M. Konuma. Plasma Techniques for Film Deposition. Alpha Science International,Harrow, U.K., 2005.
[4] F.F. Chen. Introduction to Plasma Physics and Controlled Fusion, volume 1. PlenumPress, New York, 1984.
[5] A. Grill. Cold Plasma in Metarials Fabrication: from Fundamentals to Applications.Institute of Electrical and Electronics Engineers Press, Inc., Piscataway, NJ, 1994.
[6] A. Ganguli and R.D. Tarey. Unserstanding plasma sources. Current Science, 83(3):279–290, August 2002.
[7] A. Fridman and L.A. Kennedy. Plasma Physics and Engineering. Taylor and Francis,New York, 2004.
[8] M.A. Lieberman and A.J. Lichtenberg. Principles of Plasma Discharges and MetarialsProcessing, page 547. Willey, Hoboken, New Jersey, 2 edition, 2005.
[9] M.A. Lieberman and A.J. Lichtenberg. Principles of Plasma Discharges and MetarialsProcessing, page 460. Willey, New York, 1 edition, 1994.
[10] D.M. Pozar. Microwave Engineering, pages 104–141. Wiley, New York, 2 edition,1998.
[11] I.G. Brown. The Physics and Technology of Ion Sources. Willey-VCH, Weinheim, 2edition, 2004.
[12] A. Fridman and L.A. Kennedy. Handbook of Plasma Processing Technology : Fun-damentals, Etching, Deposition, and Surface Interactions. Noyes Publications, ParkRidge, N.J., U.S.A., 1990.
[13] R.H. Huddlesyone and L.L. Stanley. Plasma Diagnostic Techniques. Academic Press,New York, 1965.
[14] Yang Z. Xiao K. Jin, D. and J. Dai. Diagnosis of hydrogen plasma in a miniature penningion source by double probes. Plasma Science and Technology, 11(1):48–51, February2009.
56